{"id":29904,"date":"2024-10-28T03:18:35","date_gmt":"2024-10-28T03:18:35","guid":{"rendered":"http:\/\/atmokpo.com\/w\/?p=29904"},"modified":"2024-11-26T06:50:54","modified_gmt":"2024-11-26T06:50:54","slug":"%eb%94%a5%eb%9f%ac%eb%8b%9d-%ed%8c%8c%ec%9d%b4%ed%86%a0%ec%b9%98-%ea%b0%95%ec%a2%8c-faster-r-cnn","status":"publish","type":"post","link":"https:\/\/atmokpo.com\/w\/29904\/","title":{"rendered":"\ub525\ub7ec\ub2dd \ud30c\uc774\ud1a0\uce58 \uac15\uc88c, Faster R-CNN"},"content":{"rendered":"<article>\n<p>\ubcf8 \uac15\uc88c\uc5d0\uc11c\ub294 \ub525\ub7ec\ub2dd\uc744 \ud65c\uc6a9\ud55c \uac1d\uccb4 \ud0d0\uc9c0(Object Detection) \uae30\ubc95 \uc911 \ud558\ub098\uc778 Faster R-CNN(Region-based Convolutional Neural Network)\uc5d0 \ub300\ud574 \ub2e4\ub8f9\ub2c8\ub2e4. \ub610\ud55c, PyTorch \ud504\ub808\uc784\uc6cc\ud06c\ub97c \uc0ac\uc6a9\ud558\uc5ec Faster R-CNN\uc744 \uad6c\ud604\ud558\uace0 \uc2e4\uc81c \ub370\uc774\ud130\ub97c \ud1b5\ud574 \ud559\uc2b5\uc2dc\ud0a4\ub294 \uacfc\uc815\uc744 \uc124\uba85\ud558\uaca0\uc2b5\ub2c8\ub2e4.<\/p>\n<h2>1. Faster R-CNN \uac1c\uc694<\/h2>\n<p>Faster R-CNN\uc740 \uc774\ubbf8\uc9c0 \ub0b4 \uac1d\uccb4\ub97c \ud0d0\uc9c0\ud558\ub294 \ub370 \uc788\uc5b4 \ub192\uc740 \uc815\ud655\ub3c4\ub97c \uc790\ub791\ud558\ub294 \ub525\ub7ec\ub2dd \ubaa8\ub378\uc785\ub2c8\ub2e4. \uc774\ub294 CNN(Convolutional Neural Network) \uae30\ubc18\uc758 \ub450 \uac00\uc9c0 \uc8fc\uc694 \uad6c\uc131 \uc694\uc18c\uc778:<\/p>\n<ul>\n<li><strong>Region Proposal Network (RPN)<\/strong>: \uc7a0\uc7ac\uc801\uc778 \uac1d\uccb4 \uc601\uc5ed\uc744 \uc81c\uc548\ud558\ub294 \uc5ed\ud560\uc744 \ud569\ub2c8\ub2e4.<\/li>\n<li><strong>Fast R-CNN<\/strong>: RPN\uc5d0\uc11c \ub098\uc628 \uc601\uc5ed\ub4e4\uc744 \uc815\uc81c\ud558\uc5ec \ucd5c\uc885 \uac1d\uccb4 \ud074\ub798\uc2a4\uc640 \ubc14\uc6b4\ub529 \ubc15\uc2a4\ub97c \uc608\uce21\ud569\ub2c8\ub2e4.<\/li>\n<\/ul>\n<p>Faster R-CNN\uc758 \uc8fc\uc694 \uac15\uc810\uc740 RPN\uc774 CNN\uc758 \uae30\uc6b8\uae30\ub97c \uc9c1\uc811 \uacf5\uc720\ud558\uc5ec \uac1d\uccb4 \uc81c\uc548\uc744 \uc218\ud589\ud558\uae30 \ub54c\ubb38\uc5d0 \uc774\uc804\uc758 \ubc29\ubc95\ub4e4\ubcf4\ub2e4 \ud6e8\uc52c \ube60\ub974\uace0 \ud6a8\uc728\uc801\uc774\ub77c\ub294 \uc810\uc785\ub2c8\ub2e4.<\/p>\n<h2>2. Faster R-CNN\uc758 \uc791\ub3d9 \uc6d0\ub9ac<\/h2>\n<p>Faster R-CNN\uc740 \ub2e4\uc74c\uacfc \uac19\uc740 \ub2e8\uacc4\ub85c \uc791\ub3d9\ud569\ub2c8\ub2e4:<\/p>\n<ol>\n<li>\uc785\ub825 \uc774\ubbf8\uc9c0\ub97c CNN\uc5d0 \ud1b5\uacfc\uc2dc\ucf1c \ud53c\uccd0 \ub9f5(feature map)\uc744 \uc0dd\uc131\ud569\ub2c8\ub2e4.<\/li>\n<li>\ud53c\uccd0 \ub9f5\uc744 \uae30\ubc18\uc73c\ub85c RPN\uc774 \uc81c\uc548\ub41c \uac1d\uccb4 \uc601\uc5ed(Region Proposals)\uc744 \uc0dd\uc131\ud569\ub2c8\ub2e4.<\/li>\n<li>\uc81c\uc548\ub41c \uc601\uc5ed\uc744 \uae30\ubc18\uc73c\ub85c Fast R-CNN\uc774 \uac01 \uc601\uc5ed\uc758 \ud074\ub798\uc2a4\ub97c \uc608\uce21\ud558\uace0, \ubc14\uc6b4\ub529 \ubc15\uc2a4\ub97c \ubcf4\uc815\ud569\ub2c8\ub2e4.<\/li>\n<\/ol>\n<p>\uc774\ub7ec\ud55c \ubd80\ubd84\ub4e4\uc740 \ubaa8\ub450 \ud559\uc2b5 \uacfc\uc815\uc5d0\uc11c \ud30c\ub77c\ubbf8\ud130\ub97c \uc870\uc815\ud558\uae30 \ub54c\ubb38\uc5d0, \ub370\uc774\ud130\uac00 \uc798 \uc900\ube44\ub418\uc5b4 \uc788\ub2e4\uba74 \ub192\uc740 \uc131\ub2a5\uc744 \uc774\ub04c\uc5b4\ub0bc \uc218 \uc788\uc2b5\ub2c8\ub2e4.<\/p>\n<h2>3. \ud658\uacbd \uc124\uc815<\/h2>\n<p>Faster R-CNN\uc744 \uad6c\ud604\ud558\uae30 \uc704\ud574 \ud544\uc694\ud55c \ub77c\uc774\ube0c\ub7ec\ub9ac\ub294 \ub2e4\uc74c\uacfc \uac19\uc2b5\ub2c8\ub2e4:<\/p>\n<ul>\n<li><strong>torch<\/strong>: PyTorch \ub77c\uc774\ube0c\ub7ec\ub9ac<\/li>\n<li><strong>torchvision<\/strong>: \uc774\ubbf8\uc9c0 \ucc98\ub9ac \ubc0f \uc804\ucc98\ub9ac \uae30\ub2a5\uc744 \uc81c\uacf5<\/li>\n<li><strong>numpy<\/strong>: \ubc30\uc5f4 \ubc0f \uc218\uce58 \uacc4\uc0b0\uc5d0 \ud544\uc694\ud55c \ub77c\uc774\ube0c\ub7ec\ub9ac<\/li>\n<li><strong>matplotlib<\/strong>: \uacb0\uacfc \uc2dc\uac01\ud654\uc5d0 \uc0ac\uc6a9<\/li>\n<\/ul>\n<p>\uc774 \uc678\uc5d0\ub3c4 \ub370\uc774\ud130\uc14b\uc744 \ub2e4\ub8e8\uae30 \uc704\ud574 <strong>torchvision.datasets<\/strong>\uc5d0\uc11c \uc81c\uacf5\ud558\ub294 \ub370\uc774\ud130\uc14b\uc744 \ud65c\uc6a9\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.<\/p>\n<h3>3.1. \ub77c\uc774\ube0c\ub7ec\ub9ac \uc124\uce58<\/h3>\n<p>\uc544\ub798 \ucf54\ub4dc\ub97c \ud1b5\ud574 \ud544\uc694 \ub77c\uc774\ube0c\ub7ec\ub9ac\ub97c \uc124\uce58\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4:<\/p>\n<pre><code>pip install torch torchvision numpy matplotlib<\/code><\/pre>\n<h2>4. \ub370\uc774\ud130\uc14b \uc900\ube44<\/h2>\n<p>Faster R-CNN\uc744 \ud559\uc2b5\uc2dc\ud0a4\uae30 \uc704\ud574 \uc0ac\uc6a9\ud560 \ub370\uc774\ud130\uc14b\uc740 PASCAL VOC, COCO, \ub610\ub294 \uc9c1\uc811 \uad6c\ucd95\ud55c \ub370\uc774\ud130\uc14b\uc77c \uc218 \uc788\uc2b5\ub2c8\ub2e4. \uc5ec\uae30\uc5d0\uc11c\ub294 COCO \ub370\uc774\ud130\uc14b\uc744 \uc0ac\uc6a9\ud574 \ubcf4\uaca0\uc2b5\ub2c8\ub2e4.<\/p>\n<h3>4.1. COCO \ub370\uc774\ud130\uc14b \ub2e4\uc6b4\ub85c\ub4dc<\/h3>\n<p>COCO \ub370\uc774\ud130\uc14b\uc740 \uc5ec\ub7ec \uacf5\uac1c\ub41c \uc18c\uc2a4\uc5d0\uc11c \ub2e4\uc6b4\ub85c\ub4dc\ud560 \uc218 \uc788\uc73c\uba70, \uc774\ub97c PyTorch\uc758 Dataloader\ub97c \ud1b5\ud574 \uc27d\uac8c \ubd88\ub7ec\uc62c \uc218 \uc788\uc2b5\ub2c8\ub2e4. \ud544\uc694\ud55c \ub370\uc774\ud130\uc14b\uc740 [COCO \ub370\uc774\ud130\uc14b \uacf5\uc2dd \uc6f9\uc0ac\uc774\ud2b8](https:\/\/cocodataset.org\/#download)\uc5d0\uc11c \ub2e4\uc6b4\ub85c\ub4dc\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.<\/p>\n<h2>5. Faster R-CNN \ubaa8\ub378 \uad6c\ud604<\/h2>\n<p>\uc774\uc81c PyTorch\ub85c Faster R-CNN \ubaa8\ub378\uc744 \uad6c\ucd95\ud574\ubcf4\uaca0\uc2b5\ub2c8\ub2e4. PyTorch\uc758 torchvision \ud328\ud0a4\uc9c0\ub97c \uc0ac\uc6a9\ud558\uba74 \uae30\ubcf8 \ud504\ub808\uc784\uc6cc\ud06c\ub97c \uc190\uc27d\uac8c \ud65c\uc6a9\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.<\/p>\n<h3>5.1. \ubaa8\ub378 \ub85c\ub4dc<\/h3>\n<p>Pre-trained \ubaa8\ub378\uc744 \ub85c\ub4dc\ud558\uc5ec Transfer Learning\uc744 \uc218\ud589\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4. \uc774\ub97c \ud1b5\ud574 \ud559\uc2b5 \uc18d\ub3c4\uc640 \uc131\ub2a5\uc744 \uac1c\uc120\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.<\/p>\n<pre><code>\nimport torch\nimport torchvision\nfrom torchvision.models.detection import FasterRCNN\nfrom torchvision.models.detection.rpn import AnchorGenerator\n\n# Pre-trained Faster R-CNN \ubaa8\ub378 \ub85c\ub4dc\nmodel = torchvision.models.detection.fasterrcnn_resnet50_fpn(pretrained=True)\n\n# \ubaa8\ub378\uc758 \ubd84\ub958\uae30\ub97c \uc870\uc815\nnum_classes = 91  # COCO \ub370\uc774\ud130\uc14b\uc758 \ud074\ub798\uc218 \uc218\nin_features = model.roi_heads.box_predictor.cls_score.in_features\nmodel.roi_heads.box_predictor = torchvision.models.detection.faster_rcnn.FastRCNNPredictor(in_features, num_classes)\n    <\/code><\/pre>\n<h3>5.2. \ub370\uc774\ud130 \uc804\ucc98\ub9ac<\/h3>\n<p>\ub370\uc774\ud130\ub97c \ubaa8\ub378\uc5d0 \ub9de\ucdb0 \uc804\ucc98\ub9ac\ud558\ub294 \uacfc\uc815\uc774 \ud544\uc694\ud569\ub2c8\ub2e4. \uc774\ubbf8\uc9c0\ub97c \ud150\uc11c\ub85c \ubcc0\ud658\ud558\uace0, \uc815\uaddc\ud654 \uacfc\uc815\uc744 \uc218\ud589\ud574\uc57c \ud569\ub2c8\ub2e4.<\/p>\n<pre><code>\nfrom torchvision import transforms\n\ntransform = transforms.Compose([\n    transforms.ToTensor(),\n    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),\n])\n    <\/code><\/pre>\n<h3>5.3. \ub370\uc774\ud130 \ub85c\ub354 \uad6c\uc131<\/h3>\n<p>PyTorch\uc758 DataLoader\ub97c \uc0ac\uc6a9\ud558\uc5ec \ubc30\uce58 \ub2e8\uc704\ub85c \ub370\uc774\ud130\ub97c \ud6a8\uc728\uc801\uc73c\ub85c \ub85c\ub4dc\ud569\ub2c8\ub2e4.<\/p>\n<pre><code>\nfrom torch.utils.data import DataLoader\nfrom torchvision.datasets import CocoDetection\n\ndataset = CocoDetection(root='path\/to\/coco\/train2017',\n                         annFile='path\/to\/coco\/annotations\/instances_train2017.json',\n                         transform=transform)\n\ndata_loader = DataLoader(dataset, batch_size=4, shuffle=True, collate_fn=lambda x: tuple(zip(*x)))\n    <\/code><\/pre>\n<h2>6. \ubaa8\ub378 \ud6c8\ub828<\/h2>\n<p>\uc774\uc81c \ubaa8\ub378\uc744 \ud6c8\ub828\ud560 \uc900\ube44\uac00 \ub418\uc5c8\uc2b5\ub2c8\ub2e4. \uc635\ud2f0\ub9c8\uc774\uc800\uc640 \uc190\uc2e4 \ud568\uc218\ub97c \uc815\uc758\ud558\uace0, \uc5d0\ud3ed\uc744 \uc124\uc815\ud558\uc5ec \ubaa8\ub378\uc744 \ud559\uc2b5\uc2dc\ud0b5\ub2c8\ub2e4.<\/p>\n<h3>6.1. \uc190\uc2e4 \ubc0f \uc635\ud2f0\ub9c8\uc774\uc800 \uc815\uc758<\/h3>\n<pre><code>\ndevice = torch.device('cuda') if torch.cuda.is_available() else torch.device('cpu')\nmodel.to(device)\n\nparams = [p for p in model.parameters() if p.requires_grad]\noptimizer = torch.optim.SGD(params, lr=0.005, momentum=0.9, weight_decay=0.0005)\n    <\/code><\/pre>\n<h3>6.2. \ud6c8\ub828 \ub8e8\ud504<\/h3>\n<pre><code>\nnum_epochs = 10\n\nfor epoch in range(num_epochs):\n    model.train()\n    for images, targets in data_loader:\n        images = list(image.to(device) for image in images)\n        targets = [{k: v.to(device) for k, v in t.items()} for t in targets]\n    \n        # \uae30\uc6b8\uae30 \ucd08\uae30\ud654\n        optimizer.zero_grad()\n    \n        # \ubaa8\ub378\uc758 \uc608\uce21 \uacb0\uacfc\n        loss_dict = model(images, targets)\n    \n        # \uc190\uc2e4 \uacc4\uc0b0\n        losses = sum(loss for loss in loss_dict.values())\n    \n        # \uc5ed\uc804\ud30c\n        losses.backward()\n        optimizer.step()\n    \n    print(f\"Epoch {epoch+1}\/{num_epochs}, Loss: {losses.item()}\")\n    <\/code><\/pre>\n<h2>7. \uac80\uc99d \ubc0f \ud3c9\uac00<\/h2>\n<p>\ubaa8\ub378\uc758 \uc131\ub2a5\uc744 \ud3c9\uac00\ud558\uae30 \uc704\ud574 \uac80\uc99d \ub370\uc774\ud130\uc14b\uc744 \uc0ac\uc6a9\ud558\uc5ec \ud559\uc2b5\ub41c \ubaa8\ub378\uc744 \ud14c\uc2a4\ud2b8\ud569\ub2c8\ub2e4.<\/p>\n<h3>7.1. \ud3c9\uac00 \ud568\uc218 \uc815\uc758<\/h3>\n<pre><code>\ndef evaluate(model, data_loader):\n    model.eval()\n    list_of_boxes = []\n    list_of_scores = []\n    list_of_labels = []\n\n    with torch.no_grad():\n        for images, targets in data_loader:\n            images = list(image.to(device) for image in images)\n            outputs = model(images)\n     \n            # \uacb0\uacfc \uc800\uc7a5\n            for output in outputs:\n                list_of_boxes.append(output['boxes'].cpu().numpy())\n                list_of_scores.append(output['scores'].cpu().numpy())\n                list_of_labels.append(output['labels'].cpu().numpy())\n\n    return list_of_boxes, list_of_scores, list_of_labels\n    <\/code><\/pre>\n<h2>8. \uacb0\uacfc \uc2dc\uac01\ud654<\/h2>\n<p>\ubaa8\ub378\uc758 \uac1d\uccb4 \ud0d0\uc9c0 \uacb0\uacfc\ub97c \uc2dc\uac01\ud654\ud558\uc5ec \uc2e4\uc81c\ub85c \uc5bc\ub9c8\ub098 \uc798 \uc791\ub3d9\ud558\ub294\uc9c0 \ud655\uc778\ud569\ub2c8\ub2e4.<\/p>\n<pre><code>\nimport matplotlib.pyplot as plt\nimport torchvision.transforms.functional as F\n\ndef visualize_results(images, boxes, labels):\n    for img, box, label in zip(images, boxes, labels):\n        img = F.to_pil_image(img)\n        plt.imshow(img)\n\n        for b, l in zip(box, label):\n            xmin, ymin, xmax, ymax = b\n            plt.gca().add_patch(plt.Rectangle((xmin, ymin), xmax - xmin, ymax - ymin,\n                                    fill=False, edgecolor='red', linewidth=3))\n            plt.text(xmin, ymin, f'Class: {l}', bbox=dict(facecolor='yellow', alpha=0.5))\n\n        plt.axis('off')\n        plt.show()\n\n# \uc774\ubbf8\uc9c0\uc640 \uc815\ub2f5\uc744 \ubd88\ub7ec\uc628 \ud6c4 \uc2dc\uac01\ud654\nimages, targets = next(iter(data_loader))\nboxes, scores, labels = evaluate(model, [images])\nvisualize_results(images, boxes, labels)\n    <\/code><\/pre>\n<h2>9. \uacb0\ub860<\/h2>\n<p>\uc774\ubc88 \uac15\uc88c\uc5d0\uc11c\ub294 Faster R-CNN\uc744 PyTorch\ub85c \uad6c\ud604\ud558\ub294 \ubc29\ubc95\uc744 \uc54c\uc544\ubcf4\uc558\uc2b5\ub2c8\ub2e4. \uac1d\uccb4 \ud0d0\uc9c0\uc758 \uae30\ubcf8 \uc6d0\ub9ac\uc640 RPN, Fast R-CNN\uc758 \uc791\ub3d9 \ubc29\uc2dd\uc744 \uc774\ud574\ud588\uc73c\uba70, \ud559\uc2b5\uacfc \uac80\uc99d, \uc2dc\uac01\ud654 \uacfc\uc815\uc744 \ud1b5\ud574 \ubaa8\ub378\uc758 \uc131\ub2a5\uc744 \ud655\uc778\ud560 \uc218 \uc788\uc5c8\uc2b5\ub2c8\ub2e4. \uc2e4\uc81c \ud504\ub85c\uc81d\ud2b8\uc5d0 \ud65c\uc6a9\ud558\uc5ec \uc5ec\ub7ec\ubd84\uc758 \ub370\uc774\ud130\uc5d0 \ub9de\ub294 \uac1d\uccb4 \ud0d0\uc9c0 \ubaa8\ub378\uc744 \uad6c\ucd95\ud574 \ubcf4\uc2dc\uae30\ub97c \ubc14\ub78d\ub2c8\ub2e4.<\/p>\n<\/article>\n","protected":false},"excerpt":{"rendered":"<p>\ubcf8 \uac15\uc88c\uc5d0\uc11c\ub294 \ub525\ub7ec\ub2dd\uc744 \ud65c\uc6a9\ud55c \uac1d\uccb4 \ud0d0\uc9c0(Object Detection) \uae30\ubc95 \uc911 \ud558\ub098\uc778 Faster R-CNN(Region-based Convolutional Neural Network)\uc5d0 \ub300\ud574 \ub2e4\ub8f9\ub2c8\ub2e4. \ub610\ud55c, PyTorch \ud504\ub808\uc784\uc6cc\ud06c\ub97c \uc0ac\uc6a9\ud558\uc5ec Faster R-CNN\uc744 \uad6c\ud604\ud558\uace0 \uc2e4\uc81c \ub370\uc774\ud130\ub97c \ud1b5\ud574 \ud559\uc2b5\uc2dc\ud0a4\ub294 \uacfc\uc815\uc744 \uc124\uba85\ud558\uaca0\uc2b5\ub2c8\ub2e4. 1. Faster R-CNN \uac1c\uc694 Faster R-CNN\uc740 \uc774\ubbf8\uc9c0 \ub0b4 \uac1d\uccb4\ub97c \ud0d0\uc9c0\ud558\ub294 \ub370 \uc788\uc5b4 \ub192\uc740 \uc815\ud655\ub3c4\ub97c \uc790\ub791\ud558\ub294 \ub525\ub7ec\ub2dd \ubaa8\ub378\uc785\ub2c8\ub2e4. \uc774\ub294 CNN(Convolutional Neural Network) \uae30\ubc18\uc758 \ub450 \uac00\uc9c0 \uc8fc\uc694 &hellip; <a href=\"https:\/\/atmokpo.com\/w\/29904\/\" class=\"more-link\">\ub354 \ubcf4\uae30<span class=\"screen-reader-text\"> &#8220;\ub525\ub7ec\ub2dd \ud30c\uc774\ud1a0\uce58 \uac15\uc88c, Faster R-CNN&#8221;<\/span><\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_jetpack_memberships_contains_paid_content":false,"footnotes":""},"categories":[33],"tags":[],"class_list":["post-29904","post","type-post","status-publish","format-standard","hentry","category-33"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v26.2 - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>\ub525\ub7ec\ub2dd \ud30c\uc774\ud1a0\uce58 \uac15\uc88c, Faster R-CNN - \ub77c\uc774\ube0c\uc2a4\ub9c8\ud2b8<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/atmokpo.com\/w\/29904\/\" \/>\n<meta property=\"og:locale\" content=\"ko_KR\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"\ub525\ub7ec\ub2dd \ud30c\uc774\ud1a0\uce58 \uac15\uc88c, Faster R-CNN - \ub77c\uc774\ube0c\uc2a4\ub9c8\ud2b8\" \/>\n<meta property=\"og:description\" content=\"\ubcf8 \uac15\uc88c\uc5d0\uc11c\ub294 \ub525\ub7ec\ub2dd\uc744 \ud65c\uc6a9\ud55c \uac1d\uccb4 \ud0d0\uc9c0(Object Detection) \uae30\ubc95 \uc911 \ud558\ub098\uc778 Faster R-CNN(Region-based Convolutional Neural Network)\uc5d0 \ub300\ud574 \ub2e4\ub8f9\ub2c8\ub2e4. \ub610\ud55c, PyTorch \ud504\ub808\uc784\uc6cc\ud06c\ub97c \uc0ac\uc6a9\ud558\uc5ec Faster R-CNN\uc744 \uad6c\ud604\ud558\uace0 \uc2e4\uc81c \ub370\uc774\ud130\ub97c \ud1b5\ud574 \ud559\uc2b5\uc2dc\ud0a4\ub294 \uacfc\uc815\uc744 \uc124\uba85\ud558\uaca0\uc2b5\ub2c8\ub2e4. 1. Faster R-CNN \uac1c\uc694 Faster R-CNN\uc740 \uc774\ubbf8\uc9c0 \ub0b4 \uac1d\uccb4\ub97c \ud0d0\uc9c0\ud558\ub294 \ub370 \uc788\uc5b4 \ub192\uc740 \uc815\ud655\ub3c4\ub97c \uc790\ub791\ud558\ub294 \ub525\ub7ec\ub2dd \ubaa8\ub378\uc785\ub2c8\ub2e4. \uc774\ub294 CNN(Convolutional Neural Network) \uae30\ubc18\uc758 \ub450 \uac00\uc9c0 \uc8fc\uc694 &hellip; \ub354 \ubcf4\uae30 &quot;\ub525\ub7ec\ub2dd \ud30c\uc774\ud1a0\uce58 \uac15\uc88c, Faster R-CNN&quot;\" \/>\n<meta property=\"og:url\" content=\"https:\/\/atmokpo.com\/w\/29904\/\" \/>\n<meta property=\"og:site_name\" content=\"\ub77c\uc774\ube0c\uc2a4\ub9c8\ud2b8\" \/>\n<meta property=\"article:published_time\" content=\"2024-10-28T03:18:35+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2024-11-26T06:50:54+00:00\" \/>\n<meta name=\"author\" content=\"root\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@bebubo4\" \/>\n<meta name=\"twitter:site\" content=\"@bebubo4\" \/>\n<meta name=\"twitter:label1\" content=\"\uae00\uc4f4\uc774\" \/>\n\t<meta name=\"twitter:data1\" content=\"root\" \/>\n\t<meta name=\"twitter:label2\" content=\"\uc608\uc0c1 \ub418\ub294 \ud310\ub3c5 \uc2dc\uac04\" \/>\n\t<meta name=\"twitter:data2\" content=\"2\ubd84\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\/\/atmokpo.com\/w\/29904\/#article\",\"isPartOf\":{\"@id\":\"https:\/\/atmokpo.com\/w\/29904\/\"},\"author\":{\"name\":\"root\",\"@id\":\"https:\/\/atmokpo.com\/w\/#\/schema\/person\/91b6b3b138fbba0efb4ae64b1abd81d7\"},\"headline\":\"\ub525\ub7ec\ub2dd \ud30c\uc774\ud1a0\uce58 \uac15\uc88c, Faster R-CNN\",\"datePublished\":\"2024-10-28T03:18:35+00:00\",\"dateModified\":\"2024-11-26T06:50:54+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\/\/atmokpo.com\/w\/29904\/\"},\"wordCount\":86,\"publisher\":{\"@id\":\"https:\/\/atmokpo.com\/w\/#organization\"},\"articleSection\":[\"\ud30c\uc774\ud1a0\uce58 \uac15\uc88c\"],\"inLanguage\":\"ko-KR\"},{\"@type\":\"WebPage\",\"@id\":\"https:\/\/atmokpo.com\/w\/29904\/\",\"url\":\"https:\/\/atmokpo.com\/w\/29904\/\",\"name\":\"\ub525\ub7ec\ub2dd \ud30c\uc774\ud1a0\uce58 \uac15\uc88c, Faster R-CNN - \ub77c\uc774\ube0c\uc2a4\ub9c8\ud2b8\",\"isPartOf\":{\"@id\":\"https:\/\/atmokpo.com\/w\/#website\"},\"datePublished\":\"2024-10-28T03:18:35+00:00\",\"dateModified\":\"2024-11-26T06:50:54+00:00\",\"breadcrumb\":{\"@id\":\"https:\/\/atmokpo.com\/w\/29904\/#breadcrumb\"},\"inLanguage\":\"ko-KR\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/atmokpo.com\/w\/29904\/\"]}]},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/atmokpo.com\/w\/29904\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"\ud648\",\"item\":\"https:\/\/atmokpo.com\/w\/en\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"\ub525\ub7ec\ub2dd \ud30c\uc774\ud1a0\uce58 \uac15\uc88c, Faster R-CNN\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/atmokpo.com\/w\/#website\",\"url\":\"https:\/\/atmokpo.com\/w\/\",\"name\":\"\ub77c\uc774\ube0c\uc2a4\ub9c8\ud2b8\",\"description\":\"\",\"publisher\":{\"@id\":\"https:\/\/atmokpo.com\/w\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/atmokpo.com\/w\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"ko-KR\"},{\"@type\":\"Organization\",\"@id\":\"https:\/\/atmokpo.com\/w\/#organization\",\"name\":\"\ub77c\uc774\ube0c\uc2a4\ub9c8\ud2b8\",\"url\":\"https:\/\/atmokpo.com\/w\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"ko-KR\",\"@id\":\"https:\/\/atmokpo.com\/w\/#\/schema\/logo\/image\/\",\"url\":\"https:\/\/atmokpo.com\/w\/wp-content\/uploads\/2024\/11\/logo.png\",\"contentUrl\":\"https:\/\/atmokpo.com\/w\/wp-content\/uploads\/2024\/11\/logo.png\",\"width\":400,\"height\":400,\"caption\":\"\ub77c\uc774\ube0c\uc2a4\ub9c8\ud2b8\"},\"image\":{\"@id\":\"https:\/\/atmokpo.com\/w\/#\/schema\/logo\/image\/\"},\"sameAs\":[\"https:\/\/x.com\/bebubo4\"]},{\"@type\":\"Person\",\"@id\":\"https:\/\/atmokpo.com\/w\/#\/schema\/person\/91b6b3b138fbba0efb4ae64b1abd81d7\",\"name\":\"root\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"ko-KR\",\"@id\":\"https:\/\/atmokpo.com\/w\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/secure.gravatar.com\/avatar\/708197b41fc6435a7ce22d951b25d4a47e9e904270cb1f04682d4f025066f80c?s=96&d=mm&r=g\",\"contentUrl\":\"https:\/\/secure.gravatar.com\/avatar\/708197b41fc6435a7ce22d951b25d4a47e9e904270cb1f04682d4f025066f80c?s=96&d=mm&r=g\",\"caption\":\"root\"},\"sameAs\":[\"http:\/\/atmokpo.com\/w\"],\"url\":\"https:\/\/atmokpo.com\/w\/author\/root\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"\ub525\ub7ec\ub2dd \ud30c\uc774\ud1a0\uce58 \uac15\uc88c, Faster R-CNN - \ub77c\uc774\ube0c\uc2a4\ub9c8\ud2b8","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/atmokpo.com\/w\/29904\/","og_locale":"ko_KR","og_type":"article","og_title":"\ub525\ub7ec\ub2dd \ud30c\uc774\ud1a0\uce58 \uac15\uc88c, Faster R-CNN - \ub77c\uc774\ube0c\uc2a4\ub9c8\ud2b8","og_description":"\ubcf8 \uac15\uc88c\uc5d0\uc11c\ub294 \ub525\ub7ec\ub2dd\uc744 \ud65c\uc6a9\ud55c \uac1d\uccb4 \ud0d0\uc9c0(Object Detection) \uae30\ubc95 \uc911 \ud558\ub098\uc778 Faster R-CNN(Region-based Convolutional Neural Network)\uc5d0 \ub300\ud574 \ub2e4\ub8f9\ub2c8\ub2e4. \ub610\ud55c, PyTorch \ud504\ub808\uc784\uc6cc\ud06c\ub97c \uc0ac\uc6a9\ud558\uc5ec Faster R-CNN\uc744 \uad6c\ud604\ud558\uace0 \uc2e4\uc81c \ub370\uc774\ud130\ub97c \ud1b5\ud574 \ud559\uc2b5\uc2dc\ud0a4\ub294 \uacfc\uc815\uc744 \uc124\uba85\ud558\uaca0\uc2b5\ub2c8\ub2e4. 1. Faster R-CNN \uac1c\uc694 Faster R-CNN\uc740 \uc774\ubbf8\uc9c0 \ub0b4 \uac1d\uccb4\ub97c \ud0d0\uc9c0\ud558\ub294 \ub370 \uc788\uc5b4 \ub192\uc740 \uc815\ud655\ub3c4\ub97c \uc790\ub791\ud558\ub294 \ub525\ub7ec\ub2dd \ubaa8\ub378\uc785\ub2c8\ub2e4. \uc774\ub294 CNN(Convolutional Neural Network) \uae30\ubc18\uc758 \ub450 \uac00\uc9c0 \uc8fc\uc694 &hellip; \ub354 \ubcf4\uae30 \"\ub525\ub7ec\ub2dd \ud30c\uc774\ud1a0\uce58 \uac15\uc88c, Faster R-CNN\"","og_url":"https:\/\/atmokpo.com\/w\/29904\/","og_site_name":"\ub77c\uc774\ube0c\uc2a4\ub9c8\ud2b8","article_published_time":"2024-10-28T03:18:35+00:00","article_modified_time":"2024-11-26T06:50:54+00:00","author":"root","twitter_card":"summary_large_image","twitter_creator":"@bebubo4","twitter_site":"@bebubo4","twitter_misc":{"\uae00\uc4f4\uc774":"root","\uc608\uc0c1 \ub418\ub294 \ud310\ub3c5 \uc2dc\uac04":"2\ubd84"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/atmokpo.com\/w\/29904\/#article","isPartOf":{"@id":"https:\/\/atmokpo.com\/w\/29904\/"},"author":{"name":"root","@id":"https:\/\/atmokpo.com\/w\/#\/schema\/person\/91b6b3b138fbba0efb4ae64b1abd81d7"},"headline":"\ub525\ub7ec\ub2dd \ud30c\uc774\ud1a0\uce58 \uac15\uc88c, Faster R-CNN","datePublished":"2024-10-28T03:18:35+00:00","dateModified":"2024-11-26T06:50:54+00:00","mainEntityOfPage":{"@id":"https:\/\/atmokpo.com\/w\/29904\/"},"wordCount":86,"publisher":{"@id":"https:\/\/atmokpo.com\/w\/#organization"},"articleSection":["\ud30c\uc774\ud1a0\uce58 \uac15\uc88c"],"inLanguage":"ko-KR"},{"@type":"WebPage","@id":"https:\/\/atmokpo.com\/w\/29904\/","url":"https:\/\/atmokpo.com\/w\/29904\/","name":"\ub525\ub7ec\ub2dd \ud30c\uc774\ud1a0\uce58 \uac15\uc88c, Faster R-CNN - \ub77c\uc774\ube0c\uc2a4\ub9c8\ud2b8","isPartOf":{"@id":"https:\/\/atmokpo.com\/w\/#website"},"datePublished":"2024-10-28T03:18:35+00:00","dateModified":"2024-11-26T06:50:54+00:00","breadcrumb":{"@id":"https:\/\/atmokpo.com\/w\/29904\/#breadcrumb"},"inLanguage":"ko-KR","potentialAction":[{"@type":"ReadAction","target":["https:\/\/atmokpo.com\/w\/29904\/"]}]},{"@type":"BreadcrumbList","@id":"https:\/\/atmokpo.com\/w\/29904\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"\ud648","item":"https:\/\/atmokpo.com\/w\/en\/"},{"@type":"ListItem","position":2,"name":"\ub525\ub7ec\ub2dd \ud30c\uc774\ud1a0\uce58 \uac15\uc88c, Faster R-CNN"}]},{"@type":"WebSite","@id":"https:\/\/atmokpo.com\/w\/#website","url":"https:\/\/atmokpo.com\/w\/","name":"\ub77c\uc774\ube0c\uc2a4\ub9c8\ud2b8","description":"","publisher":{"@id":"https:\/\/atmokpo.com\/w\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/atmokpo.com\/w\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"ko-KR"},{"@type":"Organization","@id":"https:\/\/atmokpo.com\/w\/#organization","name":"\ub77c\uc774\ube0c\uc2a4\ub9c8\ud2b8","url":"https:\/\/atmokpo.com\/w\/","logo":{"@type":"ImageObject","inLanguage":"ko-KR","@id":"https:\/\/atmokpo.com\/w\/#\/schema\/logo\/image\/","url":"https:\/\/atmokpo.com\/w\/wp-content\/uploads\/2024\/11\/logo.png","contentUrl":"https:\/\/atmokpo.com\/w\/wp-content\/uploads\/2024\/11\/logo.png","width":400,"height":400,"caption":"\ub77c\uc774\ube0c\uc2a4\ub9c8\ud2b8"},"image":{"@id":"https:\/\/atmokpo.com\/w\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/x.com\/bebubo4"]},{"@type":"Person","@id":"https:\/\/atmokpo.com\/w\/#\/schema\/person\/91b6b3b138fbba0efb4ae64b1abd81d7","name":"root","image":{"@type":"ImageObject","inLanguage":"ko-KR","@id":"https:\/\/atmokpo.com\/w\/#\/schema\/person\/image\/","url":"https:\/\/secure.gravatar.com\/avatar\/708197b41fc6435a7ce22d951b25d4a47e9e904270cb1f04682d4f025066f80c?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/708197b41fc6435a7ce22d951b25d4a47e9e904270cb1f04682d4f025066f80c?s=96&d=mm&r=g","caption":"root"},"sameAs":["http:\/\/atmokpo.com\/w"],"url":"https:\/\/atmokpo.com\/w\/author\/root\/"}]}},"jetpack_featured_media_url":"","jetpack_sharing_enabled":true,"jetpack-related-posts":[],"_links":{"self":[{"href":"https:\/\/atmokpo.com\/w\/wp-json\/wp\/v2\/posts\/29904","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/atmokpo.com\/w\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/atmokpo.com\/w\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/atmokpo.com\/w\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/atmokpo.com\/w\/wp-json\/wp\/v2\/comments?post=29904"}],"version-history":[{"count":1,"href":"https:\/\/atmokpo.com\/w\/wp-json\/wp\/v2\/posts\/29904\/revisions"}],"predecessor-version":[{"id":29905,"href":"https:\/\/atmokpo.com\/w\/wp-json\/wp\/v2\/posts\/29904\/revisions\/29905"}],"wp:attachment":[{"href":"https:\/\/atmokpo.com\/w\/wp-json\/wp\/v2\/media?parent=29904"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/atmokpo.com\/w\/wp-json\/wp\/v2\/categories?post=29904"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/atmokpo.com\/w\/wp-json\/wp\/v2\/tags?post=29904"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}